Absence of an Intron Splicing Silencer in Porcine Smn1 Intron 7 Confers Immunity to the Exon Skipping Mutation in Human SMN2

et al. (2014) Absence of an Intron Splicing Silencer in Porcine Smn1 Intron 7 Confers Immunity to the Exon Skipping Mutation in Human SMN2. PLoS ONE 9(6): e98841. doi:10.1371/journal.pone.0098841 Absence of an Intron Splicing Silencer in Porcine Smn1 Intron 7 Confers Immunity to the Exon Skipping Mutation in Human SMN2 Thomas Koed Doktor 0 Lisbeth Dahl Schrder 0 Henriette Skovgaard Andersen 0 Sabrina Brner 0 Anna Kitewska 0 Charlotte Brandt Srensen 0 Brage Storstein Andresen 0 Emanuele Buratti, International Centre for Genetic Engineering and Biotechnology, Italy 0 1 Department of Biochemistry and Molecular Biology, University of Southern Denmark , Odense M, Denmark , 2 Department of Biomedicine, Aarhus University , Aarhus C, Denmark , 3 Institute of Animal Reproduction and Food Research, Polish Academy of Sciences , Olsztyn , Poland Spinal Muscular Atrophy is caused by homozygous loss of SMN1. All patients retain at least one copy of SMN2 which produces an identical protein but at lower levels due to a silent mutation in exon 7 which results in predominant exclusion of the exon. Therapies targeting the splicing of SMN2 exon 7 have been in development for several years, and their efficacy has been measured using either in vitro cellular assays or in vivo small animal models such as mice. In this study we evaluated the potential for constructing a mini-pig animal model by introducing minimal changes in the endogenous porcine Smn1 gene to maintain the native genomic structure and regulation. We found that while a Smn2-like mutation can be introduced in the porcine Smn1 gene and can diminish the function of the ESE, it would not recapitulate the splicing pattern seen in human SMN2 due to absence of a functional ISS immediately downstream of exon 7. We investigated the ISS region and show here that the porcine ISS is inactive due to disruption of a proximal hnRNP A1 binding site, while a distal hnRNP A1 binding site remains functional but is unable to maintain the functionality of the ISS as a whole. - Funding: This work was supported by a grant from The Riisfort Foundation (BSA), the Lundbeck foundation (BSA) and The Danish Medical Research Council (FSS grants no. 271-07-342 and no. 11-107174) to BSA). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: The authors have declared that no competing interests exist. The Spinal Muscular Atrophies (SMA) is a phenotypically diverse but genetically very similar group, in that the diseases are all caused by homozygous loss of the SMN1 gene [1]. The disease modifier gene, SMN2, determines to some extent the phenotype of the affected individual and is unique to the hominid line [24]. As such, SMA caused by reduced amounts of SMN protein is a disease unique to humans and the study of animal models is therefore restricted to transgenic animals. Of these, mouse models have been used extensively in the past [58], but the metabolic and physiological differences between humans and mice are limiting the potential of the mouse model for evaluation of drug candidates and studying the molecular pathology of the disease in detail. Several metabolic and physiological symptoms have been described in mouse models, which are either rare or only observed in very severe human cases [9] or more likely explained by the genetic background of the particular model [10]. The pig is in many ways a better model of human biology and mini-pigs are especially good models since they grow to app. human size and weight as adults. Pigs are known to be more genetically similar to humans than mice are [11], and their metabolism as well as physiology more close to ours than the mices. For these reasons we set out to evaluate the potential of constructing a mini-pig animal model of SMA in order to facilitate improved drug candidate testing and studies of disease pathology. Furthermore, a mini-pig model would be extremely valuable in determining the potential of stem cell treatments as the central nervous system (CNS) of pigs is very similar to the human CNS. An SMA pig model would therefore be relevant as an animal model for not only SMA but also other motor neuron diseases or in the case of traumatic injury to the motor neurons in the spinal cord. The role of SMN2 in humans is unclear in the population as a whole, but in SMA patients SMN2 serves an important function as the remaining SMN expressing gene. It fails to completely compensate for the loss of SMN1 due to aberrant splicing of exon 7 which leads to the production of predominantly truncated transcripts and a corresponding decrease in the amounts of functional protein [1214]. As SMN2 is present in all SMA patients it has been extensively studied and serves as a drug target for drugs that specifically correct splicing of exon 7 and thereby increases amounts of functional SMN protein [1517]. As such, large animal models where broader effects of both early and late treatment can be carefully examined are becoming increasingly relevant. In particular, the bioavailability and therapeutic potential of drug candidates are more easily studied in animal models that more closely resemble the physiology and metabolism of humans. Transgenic models which have been generated through a knockout/knock-in approach can potentially display pathologies unrelated to the trans-gene itself, but as a consequence of gene disruption caused by the insertion. This was recently reported in the widely used Tg(SMN2)89Ahmb mouse model of SMA [10]. In order to construct a transgenic pig which resembles the human SMA genotype as closely as possible we chose to study the potential in converting the endogenous pig Smn1 to that of a human SMN2 and in the process changing as little as possible in the endogenous gene. The aberrant splicing of human SMN2 exon 7 is caused by the loss of an exonic splicing enhancer (ESE) due to a +6C.T transition in SMN2 exon 7 relative to SMN1 exon 7, leading to loss of binding of SRSF1 and increased binding of hnRNP A1 due to strengthening of pre-existing exonic splicing silencer (ESS) motifs [12,14,18,19]. In humans, the active ESE motif is altered from CAGACAA to an inactive TAGACAA motif in SMN2, but in pigs the ESE motif is only slightly altered to CAAACAA in the wild type Smn1. This poses the question of whether or not this sequence constitutes an active ESE and if a single Smn2-like +6C.T mutation in porcine Smn1 exon 7 can disrupt the function and result in a porcine Smn2-like gene. We began by sequencing the Yucatan mini-pig Smn1 gene from genomic DNA by designing primers to amplify individual exons based on publicly available data as well as larger parts of the intronic regions surrounding exon 7 which were not publicly available at the time. Additionally, we performed 59RACE and 39RACE in order to validate previous assignment of exons and UTR regions. The resulting Yucatan Smn1 gene sequence has bee (...truncated)